Fracture behavior of endodontically treated premolars restored with direct and indirect restorations: an in vitro study

Endodontically Treated Teeth (ETT) have challenges in resisting fracture against occlusal forces. This study aimed to assess the fracture resistance (FR) of endodontically treated premolar teeth restored with an innovative self-adhesive bulk-fill resin-based composite (RBC) (Stela , SDI Limited, Australia), a short-fiber-reinforced composite (SFRC) (EverX flow, GC., Tokyo, Japan), a conventional nanohybrid (RBC) (Filtek Z250 XT, 3M ESPE, St. Paul, MN, USA) , and a Computer-Aided Design/Computer-Aided Manufacturing (CAD/CAM) Lithium Disilicate (Emax, Dentsply Sirona, Bensheim, Germany) Endocrown.

For the FR test, 44 human extracted maxillary premolars were included in the study. Endodontic treatment was done, and a single operator prepared standardized mesio‑occlusal‑distal (MOD) cavities in the teeth and divided into four groups (n=11) according to the restorative material used. Group (F1): ETT restored with Stela resin-based composite, which was inserted into the cavity as a single 5 mm increment. Group (F2): ETT restored with (EverX Flow) composite,which was packed and light-cured for 20 seconds, leaving a 2 mm occlusal layer to be restored with nanohybrid composite (Filtek Z250 XT). Group (F3): ETT restored with Filtek Z250 XT resin-based composite only. Group (F4): ETT restored with (CAD/CAM Cerec MCXL) Lithium Disilicate endocrown, preparations were scanned with the Omnicam, and restorations were designed using CEREC Premium Software (version 4.4). For direct restoration groups the proximal walls were reconstructed by using a tofflemire matrix retainer and band. All restored teeth were subjected to FR testing using a universal testing machine ( Bluehill Lite Software on an Instron® / Each specimen’s failure load was recorded in Newton). Each specimen was individually mounted on the computer-controlled machine equipped with a 5 kN load cell, with the acrylic block secured to the lower fixed head.

Statistical analyses were conducted using SPSS 27®, GraphPad Prism®, and Microsoft Excel 2016. Data normality was assessed with the Shapiro-Wilk and Kolmogorov-Smirnov tests, which indicated a nonparametric distribution. Comparison between groups revealed a significant difference (P<0.0001), with Group F4 (Endocrown) demonstrating the highest performance (1427.73 ± 156.96), compared to all direct restoration materials. In contrast, there were no significant differences among the three direct restoration groups despite slight differences in their means. Group F2 (EverX+Z250, mean: 619.87 ± 186.41) was higher than Group F1 (Stella, mean: 561.75 ± 173.74) with P = 1.000, and higher than Group F3 (Z250, mean: 452.22 ± 202.72) with P = 0.471. Group F1 (Stella) exceeded Group F3 (Z250), but this difference was not statistically significant (P = 0.838).

While lithium disilicate endocrowns exhibit superior fracture resistance, direct restorative materials may serve as a clinically acceptable alternative for routine applications. Nevertheless, these findings should be interpreted with caution, given the inherent limitations of the in vitro study design.